We present KnIT, the Knowledge Integration Toolkit, a system for accelerating scientific discovery and predicting previously unknown protein-protein interactions. Such predictions enrich biological research and are pertinent to drug discovery and the understanding of disease. Unlike a prior study, KnIT is now fully automated and demonstrably scalable. It extracts information from the scientific literature, automatically identifying direct and indirect references to protein interactions, which is knowledge that can be represented in network form. It then reasons over this network with techniques such as matrix factorization and graph diffusion to predict new, previously unknown interactions. The accuracy and scope of KnIT's knowledge extractions are validated using comparisons to structured, manually curated data sources as well as by performing retrospective studies that predict subsequent literature discoveries using literature available prior to a given date. The KnIT methodology is a step towards automated hypothesis generation from text, with potential application to other scientific domains.

Keeping up with the ever-expanding flow of data and publications is untenable and poses a fundamental bottleneck to scientific progress. Current search technologies typically find many relevant documents, but they do not extract and organize the information content of these documents or suggest new scientific hypotheses based on this organized content. We present an initial case study on KnIT, a prototype system that mines the information contained in the scientific literature, represents it explicitly in a queriable network, and then further reasons upon these data to generate novel and experimentally testable hypotheses. KnIT combines entity detection with neighbor-text feature analysis and with graph-based diffusion of information to identify potential new properties of entities that are strongly implied by existing relationships. We discuss a successful application of our approach that mines the published literature to identify new protein kinases that phosphorylate the protein tumor suppressor p53. Retrospective analysis demonstrates the accuracy of this approach and ongoing laboratory experiments suggest that kinases identified by our system may indeed phosphorylate p53. These results establish proof of principle for automated hypothesis generation and discovery based on text mining of the scientific literature.

The automation of extracting chemical names from text has significant value to biomedical and life science research. A major barrier in this task is the difficulty of getting a sizable and good quality data to train a reliable entity extraction model. Another difficulty is the selection of informative features of chemical names, since comprehensive domain knowledge on chemistry nomenclature is required. Leveraging random text generation techniques, we explore the idea of automatically creating training sets for the task of chemical name extraction. Assuming the availability of an incomplete list of chemical names, called a dictionary, we are able to generate well-controlled, random, yet realistic chemical-like training documents. We statistically analyze the construction of chemical names based on the incomplete dictionary, and propose a series of new features, without relying on any domain knowledge. Compared to state-of-the-art models learned from manually labeled data and domain knowledge, our solution shows better or comparable results in annotating real-world data with less human effort. Moreover, we report an interesting observation about the language for chemical names. That is, both the structural and semantic components of chemical names follow a Zipfian distribution, which resembles many natural languages.

Even though queries received by traditional information retrieval systems are quite short, there are many application scenarios where long natural language queries are more effective. Further, incorporating term position information can help improve results of long queries. However, the techniques for incorporating term position information have been developed for terse queries and hence, can not be directly applied to long queries. Though there exist some methods for performing proximal search for long queries, they are not scalable due to long query response times. We describe an intuitive and simple, yet effective technique that implicitly incorporates term position information for long queries in a scalable manner. Our proposed approach achieves more than 700% faster query response times while maintaining the quality of retrieved results when compared with a state-of-the-art method for performing proximal search for very long queries.

Text and images are two major sources of information in scientific literature. Information from these two media typically reinforce and complement each other, thus simplifying the process for human to extract and comprehend information. However, machines cannot create the links or have the semantic understanding between images and text. We propose to integrate text analysis and image processing techniques to bridge the gap between the two media, and discover knowledge from the combined information sources, which would be otherwise lost by traditional single-media based mining systems. The focus is on the chemical entity extraction task because images are well known to add value to the textual content in chemical literature. Annotation of US chemical patent documents demonstrates the effectiveness of our proposal.

Patents are of crucial importance for businesses, because they provide legal protection for the invented techniques, processes or products. A patent can be held for up to 20 years. However, large maintenance fees need to be paid to keep it enforceable. If the patent is deemed not valuable, the owner may decide to abandon it by stopping paying the maintenance fees to reduce the cost. For large companies or organizations, making such decisions is difficult because too many patents need to be investigated. In this paper, we introduce the new patent mining problem of automatic patent maintenance prediction, and propose a systematic solution to analyze patents for recommending patent maintenance decision. We model the patents as a heterogeneous time-evolving information network and propose new patent features to build model for a ranked prediction on whether to maintain or abandon a patent. In addition, a network-based refinement approach is proposed to further improve the performance. We have conducted experiments on the large scale United States Patent and Trademark Office (USPTO) database which contains over four million granted patents. The results show that our technique can achieve high performance.

In this paper we introduce a new Web mining and search technique - Topic Initiator Detection (TID) on the Web. Given a topic query on the Internet and the resulting collection of time-stamped web documents which contain the query keywords, the task of TID is to automatically return which web document (or its author) initiated the topic or was the first to discuss about the topic.

The emergence of new social media such as blogs, message boards, news, and web content in general has dramatically changed the ecosystems of corporations. Consumers, non-profit organizations, and other forms of communities are extremely vocal about t

In recent years, the number of patents filed by the business enterprises in the technology industry are growing rapidly, thus providing unprecedented opportunities for knowledge discovery in patent data. One important task in this regard is to employ data mining techniques to rank patents in terms of their potential to earn money through licensing. Availability of such ranking can substantially reduce enterprise IP (Intellectual Property) management costs. Unfortunately, the existing software systems in the IP domain do not address this task directly. Through our research, we build a patent ranking software, named COA (Claim Originality Analysis) that rates a patent based on its value by measuring the recency and the impact of the important phrases that appear in the "claims" section of a patent. Experiments show that COA produces meaningful ranking when comparing it with other indirect patent evaluation metrics--citation count, patent status, and attorney's rating. In reallife settings, this tool was used by beta-testers in the IBM IP department. Lawyers found it very useful in patent rating, specifically, in highlighting potentially valuable patents in a patent cluster. In this article, we describe the ranking techniques and system architecture of COA. We also present the results that validate its effectiveness.

Corporations are extremely sensitive to issues such as brand stewardship and product reputation. Traditional brand image and reputation tracking is limited to news wires and contact centre analysis. However, with the emergence of web, Consumer Genera

The emergence of new social media such as blogs, message boards, news, and Web content in general has dramatically changed the ecosystems of corporations. Consumers, non-profit organizations, and other forms of communities are extremely vocal about their opinions and perceptions on companies and their brands on the Web. The ability to leverage such "voice of the Web" to gain consumer, brand, and market insights can be truly differentiating and valuable to todaypsilas corporations. In particular, one important form of insights can be derived from sentiment analysis on Web content. Sentiment analysis traditionally emphasizes on classification of Web comments into positive, neutral, and negative categories. This paper goes beyond sentiment classification by focusing on techniques that could detect the topics that are highly correlated with the positive and negative opinions. Such techniques, when coupled with sentiment classification, can help the business analysts to understand both the overall sentiment scope as well as the drivers behind the sentiment. In this paper, we describe our overall sentiment analysis system that consists of such sentiment analysis techniques. We then detail a novel topic detection method using point-wise mutual information and term frequency distribution. We demonstrate the effectiveness of our overall approaches via several case studies on different social media data sets.

A strong brand is a major asset to any corporation. Traditional brand image and reputation tracking is limited to news wires and contact centers analysis. However, with the emergence of Web, consumer generated media (CGM), such as blogs, news forums, message boards, and web pages/sites, is rapidly transforming the way companies analyze their brand perceptions. This paper describes the next generation COBRA (Corporate Brand and Reputation Analysis) approach to mining a wide range of CGM content to discover how the social media-based community perceives the brand. The solution processes a diverse set of structured and unstructured information and mines CGM content by generating multiple taxonomies from the data. These taxonomies are then used singly and in combination to better understand important brand characteristics and hence enhance marketing and strategic decision making. We illustrate our approach with a real-world case study involving the Kraft Foodspsila Vegemite brand.

We developed a patent analysis tool, named COA (Claim Originality Analysis) at IBM Research. The system and the methods are discussed.

Corporations are extremely sensitive to issues such as brand stewardship and product reputation. Traditional brand image and reputation tracking is limited to news wires and contact centres analysis. However, with the emergence of Web, consumer generated media (COM), such as blogs, news forums, message boards, and Web pages/sites, is rapidly becoming the "voice of the people". This paper describes a COBRA (corporate brand and reputation analysis) solution that mines a wide range of COM contents for brand and reputation analysis. The solution contains a flexible ETL (Extract, Transform, and Load) engine that processes diverse sets of structured and unstructured information, a suite of analytical capabilities that mines COM content to extract semantic entities and insights out of the data, and an alerting mechanism that utilizes the analytics results to accurately generate brand and reputation alerts. We use a real-world case study to demonstrate the effectiveness of our approach.

Taxonomies are meaningful hierarchical categorizations of documents into topics reflecting the natural relationships between the documents and their business objectives. Improving the quality of these taxonomies and reducing the overall cost required to create them is an important area of research. Supervised and unsupervised text clustering are important technologies that comprise only a part of a complete solution. However, there exists a great need for the ability for a human to efficiently interact with a taxonomy during the editing and validation phase. We have developed a comprehensive approach to solving this problem, and implemented this approach in a software tool called eClassifier. eClassifier provides features to help the taxonomy editor understand and evaluate each category of a taxonomy and visualize the relationships between the categories. Multiple techniques allow the user to make changes at both the category and document level. Metrics then establish how well the resultant taxonomy can be modeled for future document classification. In this paper, we present a comprehensive set of viewing, editing and validation techniques we have implemented in the Lotus Discovery Server resulting in a significant reduction in the time required to create a quality taxonomy.

As the web and e-business have proliferated, the practice of using customer facing knowledge bases to augment customer service and support operations has increased. This can be a very efficient, scalable and cost effective way to share knowledge. The effectiveness and cost savings are proportional to the utility of the information within the knowledge base and inversely proportional to the amount of labor required in maintaining the knowledge. To address this issue, we have developed an algorithm and methodology to increase the utility of the information within a knowledge base while greatly reducing the labor required.In this paper, we describe an implementation of an algorithm and methodology for comparing a knowledge base to a set of problem tickets to determine which categories and subcategories are not well addressed within the knowledge base. We utilize text clustering on problem ticket text to determine a set of problem categories. We then compare each knowledge base solution document to each problem category centroid using a cosine distance metric. The distance between the "closest" solution document and the corresponding centroid becomes the basis of that problem category's "knowledge gap". Our claim is that this gap metric serves as a useful method for quickly and automatically determining which problem categories have no relevant solutions in a knowledge base. We have implemented our approach, and we present the results of performing a knowledge gap analysis on a set of support center problem tickets.